Dubstepforum.com

I'm guessing most people here may have heard this track:
Soundcloud

I started a remix yesterday and unintentionally started making this squeely guitar style lead and then realised it's kinda similar to the synth KTN uses at around 1:50 and 2:05. I'm struggling to get mine to sound as raw and pronounced as his - it's sounds like quite a basic sound but his just seems to tear through the mix, any tips guys?

Maybe some saturation and compression, assuming you got the base sound. Also try bumping up the unisono to extreme.

mthrfnk wrote:I'm guessing most people here may have heard this track:
Soundcloud

I started a remix yesterday and unintentionally started making this squeely guitar style lead and then realised it's kinda similar to the synth KTN uses at around 1:50 and 2:05. I'm struggling to get mine to sound as raw and pronounced as his - it's sounds like quite a basic sound but his just seems to tear through the mix, any tips guys?

Here is generally how. But lots of vibrato, and increase the vibrato as the sound goes higher.
Put the portamento sliding at really really fast.
Make notes that are hundreds of notes away from each other. Like, C-8 and C8, (16 octaves)
And have it slide up Realllyyyyyyyyyyyyyyy fast. Then save the sound into an audio wave. And cut it short before it gets too high, that way it goes up fast, and high, but you can control what note it stops at, in this case it sounds like its going from like C-1 to C3, but the slide rate is so fast, that he probably had the octaves spread apart, and chose a section of the sample and used that.

Apply tons of saturation, tube distortion, and compress the shit out of it to make it thick and chunky.
If you can, add a lot of voices, like 8-16, and add a slight bit of offtune. Unison with slight detune.
Then saturate it more and compress it more. Put a subbass layer under it, and do some MID eq boosting and notch filtering, cut out the sounds that hurt your brain, because it won't sound perfect straight off. You should have this sound pretty easy.

jujment699 wrote: Make notes that are hundreds of notes away from each other. Like, C-8 and C8, (16 octaves)

Everything you said sounds about right, but 16 octaves??? haha even the highest one is only like 5 octaves above fundamental bass note I think... also pretty sure C-8 would be like less than 1 hz so you wouldn't hear it anyway...

Why not just slide from C-1 to C3, and if you want the slide to happen faster than decrease the portamento?

Add9 wrote:

jujment699 wrote: Make notes that are hundreds of notes away from each other. Like, C-8 and C8, (16 octaves)

Everything you said sounds about right, but 16 octaves??? haha even the highest one is only like 5 octaves above fundamental bass note I think... also pretty sure C-8 would be like less than 1 hz so you wouldn't hear it anyway...

Why not just slide from C-1 to C3, and if you want the slide to happen faster than decrease the portamento?

That won't work because the curve is exponential. Its a J curve.
Go try it.
Place a note at your LOWEST note, and a note at your HIGHEST note, and set the glide to like as fast as you can go without it sounding like two seperate notes.

Then chop the bit you want.

Then copy and paste all that, but bring the notes closer together, compare that bit to the bit you had from the audio.wav, you won't have the speed that you want. In fact, the speed shouldn't even go that high. You won't be able to match it.

I've done stuff like this before, not like this song, but using this technique. Its how he gets those whiney fluxes in his guitar. The really squealy parts that happen for not even 1/10th of a second where it just hits you right there.

You'll see, its a J curve.

ravinglogs wrote:

Add9 wrote:

jujment699 wrote: Make notes that are hundreds of notes away from each other. Like, C-8 and C8, (16 octaves)

Everything you said sounds about right, but 16 octaves??? haha even the highest one is only like 5 octaves above fundamental bass note I think... also pretty sure C-8 would be like less than 1 hz so you wouldn't hear it anyway...

Why not just slide from C-1 to C3, and if you want the slide to happen faster than decrease the portamento?

That won't work because the curve is exponential. Its a J curve.
Go try it.
Place a note at your LOWEST note, and a note at your HIGHEST note, and set the glide to like as fast as you can go without it sounding like two seperate notes.

Then chop the bit you want.

Then copy and paste all that, but bring the notes closer together, compare that bit to the bit you had from the audio.wav, you won't have the speed that you want. In fact, the speed shouldn't even go that high. You won't be able to match it.

I've done stuff like this before, not like this song, but using this technique. Its how he gets those whiney fluxes in his guitar. The really squealy parts that happen for not even 1/10th of a second where it just hits you right there.

You'll see, its a J curve.

^ He gets it. It's all in the J curve :3

Add9 wrote:

jujment699 wrote: Make notes that are hundreds of notes away from each other. Like, C-8 and C8, (16 octaves)

Everything you said sounds about right, but 16 octaves??? haha even the highest one is only like 5 octaves above fundamental bass note I think... also pretty sure C-8 would be like less than 1 hz so you wouldn't hear it anyway...

Why not just slide from C-1 to C3, and if you want the slide to happen faster than decrease the portamento?

Lets try C0 to C5 (Five ocataves)
1hz doesn' t matter because it contributes to the exponential rate.
If i start at 5 and go to 10 and the exponential rate is compoud^2 (compound meaning 2*2 is 4, 4*4 is 16, 16*16 is 256, 256*256 is 65536, = Final Product)
Then 5 - 10 is only going to leave me with 4295033108 iterations.

Where as C-8 to C8 is 16 steps, and unfortunately the curve can only go through 10 (^2) logical iterations, otherwise we end up with infinity (Straight line.) So really, once you reach 10 octaves apart, you will not get any faster. Because 9 iterations is 1.340781e^154(^2). 10 is infinity (where it stops going up the J curve and just goes straight forever.) And that is a lot of speed when you finally get to about the 5th octave, its like a whoosh sound. This makes it excel at an unmeasured speed. And since we don't want the part where it wooshes and goes to inifinty in a fraction of a second, we clip out the part where it starts to really get going, probably about the sixth-seventh iteration, and we get a part of it that will allow an unnatural squeaky curve right at the end if you catch it right before it accelerates up to infinity.

Then we hold that position for the duration of the note and reverse the sample to go back down. It will sound relatively normal to the human ear, you won't notice the acceleration because its stopped before it goes off, but thats what gives it that big WEU bit, when it reaches its peak, the cutoff.

Its like making a "WHOOP" sound, the ones knife party always use.
Anyway, i tried to be as clear as i could, if you don't get it, just try it.

The way you're interpreting it is linear, straight line, straight value.

Why not just use an envelope on the pitch with an exponential attack and release section? (not trolling here I'm really curious). Or, why not just automate the pitch to make the curve however you like it (rather than go through the resampling process)?

Add9 wrote:Why not just use an envelope on the pitch with an exponential attack and release section? (not trolling here I'm really curious). Or, why not just automate the pitch to make the curve however you like it (rather than go through the resampling process)?

I was just doing the math, I'm sure you can do it anyway you like, i was just explaining the exponential and linear qualities. Maybe the other guy can answer your question for you.
But i'm sure you can do it that way. Samplers usually have pitch envelopes.

The only reason i'd suspect using it that way is more control, and also maybe more control over the type of J curve and the linear qualities. Not to mention theres only so much you can do with an envelope. It would take many envelopes to recreate said J curve. Those numbers are extremely steep. Makes for a really tight sound. Really really tight.

WolfCryOfficial wrote:

Add9 wrote:Why not just use an envelope on the pitch with an exponential attack and release section? (not trolling here I'm really curious). Or, why not just automate the pitch to make the curve however you like it (rather than go through the resampling process)?

I was just doing the math, I'm sure you can do it anyway you like, i was just explaining the exponential and linear qualities. Maybe the other guy can answer your question for you.
But i'm sure you can do it that way. Samplers usually have pitch envelopes.

The only reason i'd suspect using it that way is more control, and also maybe more control over the type of J curve and the linear qualities. Not to mention theres only so much you can do with an envelope. It would take many envelopes to recreate said J curve. Those numbers are extremely steep. Makes for a really tight sound. Really really tight.

All right I'm gonna have to try it your way then. I do want a tight sound, after all

Add9 wrote:
All right I'm gonna have to try it your way then. I do want a tight sound, after all

Haha alright, if it works out, creds to jujment699 he proposed the idea i just reworked it into number
And you're def not gonna get it first try, i'll post my results tomorrow after work, will take a lot of guess work to get the right portamento, but it seems to be a dope method.